[trafficserver] branch master updated: TS-4536 Don't always refresh an IP when serving stale (#713)

[ja...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

jacksontj pushed a commit to branch master
in repository https://git-dual.apache.org/repos/asf/trafficserver.git

The following commit(s) were added to refs/heads/master by this push:
       new  70e11a2   TS-4536 Don't always refresh an IP when serving stale (#713)
70e11a2 is described below

commit 70e11a2cf26dcf6042487d3a7a280f91608ee336
Author: Thomas Jackson <ja...@gmail.com>
AuthorDate: Tue Jun 14 12:18:10 2016 -0700

    TS-4536 Don't always refresh an IP when serving stale (#713)
    
    With this as it stands, if a record is configured to serve stale the "creation" timeout keeps getting moved. This means if requests are made every second a record would NEVER expire-- which isn't what we want.
    
    this closes #713
---
 iocore/hostdb/HostDB.cc | 1 -
 1 file changed, 1 deletion(-)

diff --git a/iocore/hostdb/HostDB.cc b/iocore/hostdb/HostDB.cc
index 33c6459..8bb5485 100644
--- a/iocore/hostdb/HostDB.cc
+++ b/iocore/hostdb/HostDB.cc
@@ -691,7 +691,6 @@ probe(ProxyMutex *mutex, HostDBMD5 const &md5, bool ignore_timeout)
       if ((!ignore_timeout && r->is_ip_stale() && !cluster_machine_at_depth(master_hash(md5.hash)) && !r->reverse_dns) ||
           (r->is_ip_timeout() && r->serve_stale_but_revalidate())) {
         Debug("hostdb", "stale %u %u %u, using it and refreshing it", r->ip_interval(), r->ip_timestamp, r->ip_timeout_interval);
-        r->refresh_ip();
         if (!is_dotted_form_hostname(md5.host_name)) {
           HostDBContinuation *c = hostDBContAllocator.alloc();
           HostDBContinuation::Options copt;

-- 
To stop receiving notification emails like this one, please contact
['"commits@trafficserver.apache.org" <co...@trafficserver.apache.org>'].